DESCRIPTION Abstract: Many proteins involved in protein-protein interactions are considered to be undruggable due to the large, flat binding surfaces involved in complex formation. Moreover, the current rules to obtain oral bioavailability and drug-like properties are not consistent with the type of molecules often needed to drug such challenging targets. This proposal outlines a plan to develop approaches for drugging proteins that are currently considered to be "undruggable." The strategy will involve quickly eliminating proteins with little chance of being druggable from fragment-based screens and developing better computational approaches to accomplish this goal. Optimized fragment-based methods, structure-based design, and improved computational tools will be developed for lead identification and optimization. To obtain molecules with good oral bioavailability, the current rules concerning druglikeness will be challenged, and the importance of molecular size, flexibility, polar surface area, charge, and other parameters will be examined systematically. In addition, assays will be developed to identify the reasons for the lack of bioavailability to determine how to correct these deficiencies. By synthesizing and testing compounds (including control compounds) in multiple assays in a systematic fashion, the guidelines and approaches for drugging challenging targets will be established. This research could revolutionize our approach to drug discovery, lead to new drugs to better treat disease in all therapeutic areas, and thus have a dramatic impact on human health. Public Health Relevance: Of the 30,000 proteins in the human genome, only about 3,000 are thought to be druggable by small organic molecules;however, many of these undruggable proteins are highly validated and could serve as important targets in drug discovery providing suitable approaches are developed to drug these technically challenging targets. Expanding the druggable genome as outlined in this proposal could a